DE102019113681A1 - Always listening and active voice assistant and vehicle operation - Google Patents

Abstract

This disclosure provides an always-listening and active voice assistant and vehicle operation. A vehicle includes a controller configured to select a group of topics for generating a response to a question embedded in the group based on an operating parameter of the vehicle and a formulation syntax. The selection is in response to an input stemming from utterances, including a previous topic and a following topic with a reference object in between that is associated with only one of the topics through the syntax. The vehicle can operate an interface to issue the answer.

Description

TECHNICAL AREA

This disclosure relates to the operation of vehicles by active and always listening voice assistance.

GENERAL PRIOR ART

Query systems answer questions asked after the call. For example, "Hey Ford®, what's the weather like today?" After calling "Hey Ford®," natural language processing and artificial methods are used to find an answer to the question. This cadence, which requires a call before the question, may require more statements than necessary to provide the answer, since conversations prior to the call are ignored. In addition, interrogation systems that provide backward cadence support may be poorly equipped to distinguish between forward cadence and reverse cadence.

SUMMARY

A vehicle includes a controller configured to select a group of topics for generating a response to a question embedded in the group based on an operating parameter of the vehicle and a formulation syntax. The selection is in response to an input stemming from utterances, including a previous topic and a following topic with a reference object in between that is associated with only one of the topics through the syntax. The vehicle can operate an interface to issue the answer.

A vehicle includes a controller configured to select a group of topics for generating a response to a question embedded in the group based on an operating parameter of the vehicle and a formulation syntax. The selection is in response to an input stemming from utterances, including a previous topic and a following topic with a reference object in between that is associated with only one of the topics through the syntax. The vehicle can be operated according to the answer.

A method includes generating a response to a question by a controller embedded in a group of topics based on a selection of one of the subjects and an operating parameter of the vehicle and a syntax, and operating the vehicle according to the answer. The rendering is in response to an input coming from utterances that define a group of phrases, including a previous topic and a following topic with a reference object in between that is associated with only one of the topics through the syntax.

list of figures

• 1 is a schematic representation of a vehicle with an infotainment system and associated communication capabilities;
• 2 is a schematic representation of vehicle control systems and peripherals;
• 3A is an algorithm of always-listening speech systems; and
• 3B is an algorithm for selecting contexts.

DETAILED DESCRIPTION

Embodiments of the present disclosure are described herein. It should be understood, however, that the disclosed embodiments are merely examples and other embodiments may take various and alternative forms. The figures are not necessarily to scale; Some features may be enlarged or reduced to show details of particular components. Thus, specific structural and functional details disclosed herein are not to be construed as limiting, but merely as a representative basis for teaching one skilled in the art to variously employ the present invention. One of ordinary skill in the art will understand that various features illustrated and described with reference to any of the figures may be combined with features illustrated in one or more other figures to produce embodiments that are not expressly illustrated or described , The illustrated combinations of features provide representative embodiments for typical applications. However, various combinations and modifications of the features consistent with the teachings of this disclosure may be desirable for particular applications or implementations.

Vehicle occupants conduct conversations in which events, circumstances or problems of the past, present and future are discussed. Rather than requiring a forward-looking call, similar to "Hey Ford®, what's the temperature outside?", An algorithm can be implemented to distinguish between forward and reverse cadence. In fact, can a vehicle interrogation system may provide responses and solutions to a vehicle based on a combination of forward cadence, reverse cadence, vehicle operating parameters, and / or algorithms that are constantly listening. Questions using a backward cadence may include a confirmation question or confirmation question formulation. The confirmation question formulation may be isolated to determine the backward cadence. The confirmation question formulation would generally follow topics requiring the response service (ie, a backward cadence), while other interrogators would generally precede topics requiring the response service (ie, forward cadence). Verbal utterances can be continuously recorded and analyzed to isolate topics and provide contexts. When the topics are organized by possible questions, the algorithm waits for a call, such as "Ford®, what do you think?". A confirmation question is any question that follows a statement instead of preceded by a statement. A confirmation question formulation may include a referral object from which the question is asked. The always-listening and answer-to-answer service provides a backward cadence, where a statement is made, the service invoked, and then the response provided. The always listening and answering service provides a forward cadence, at which the service is called, making the statement and then providing the answer. The always-listening service can also search for answers for each topic so that the answer is readily available for presentation to the occupant.

• Suzy sagt: „Das Detroit Symphonieorchester spielt heute Abend in Detroit.“ „Ford®, was denkst du?“ „Ich sage, wir haben ausreichend Kraftstoff, um nach Detroit zu kommen, jedoch wird es sehr kalt.“ (Umgekehrte Kadenz)
• Suzy sagt: „Das Detroit Symphonieorchester spielt heute Abend in Detroit.“ „Hey Ford®, ich sage, wir haben ausreichend Kraftstoff, um nach Detroit zu kommen, jedoch wird es sehr kalt.“ (vorwärts gerichtete Kadenz)

An inherent problem in providing forward and reverse response services is that a conversation may include multiple responsive topics and the subject of the occupant to be answered is not particularly clear. An exemplary conversation is provided below:

• Suzy says, "The Detroit Symphony Orchestra is playing in Detroit tonight.""Ford®, what do you think?""I say we have enough fuel to get to Detroit, but it's getting very cold." (Reverse cadence)
• Suzy says, "The Detroit Symphony Orchestra is playing in Detroit tonight.""Hey Ford®, I say we have enough fuel to get to Detroit, but it's getting very cold." (Cadence moving forward)

The previous conversation included at least three topics: 1) The Detroit Symphony Orchestra plays in Detroit tonight; 2) I say we have enough fuel to get to Detroit; and 3) it gets very cold. The topics can be isolated on the basis of syntactic, categorical and other methods. Over time, topics can be separated and isolated into specific contexts. The contexts can be large categories of topics that may require answers. The contexts can also be presented to vehicle occupants for selection. The selection may also be provided via machine learning so that topics are selected according to previously selected topics. This means that the occupant can select the topic selection that has taken place and then the machine learning algorithm would automatically update the preferred contexts. After the topic has been selected based on the context, the vehicle may provide an answer or indication to the occupant by operating the vehicle or displaying the response. In fact, an always-listening algorithm can be used to identify topics that require a response service and automatically provide a response to the topic after being called by a confirmation question formulation.

1 illustrates an exemplary system 100 that a vehicle 102 which implements an always-listening answer-polling algorithm. The vehicle 102 can a vehicle computing system (vehicle computing system - VCS ) 106 which is configured to be used over a wide area network using a telematics control unit (telematics control unit). TCU ) 120A to communicate. The TCU 120A can use different modems 122 which are configured to communicate via respective communication paths and protocols. Although in 1 an exemplary system 100 is shown, the exemplary components shown are not intended to be limiting. In fact, the system can 100 contain more or less components and additional or alternative components and / or reactions may be used.

In the vehicle 102 can be different types of automobiles, softroaders (crossover utility vehicle - CUV ), SUVs (sport utility vehicle - SUV ), Trucks, recreational vehicles (RV), boats, airplanes or other mobile machines for transporting people or goods. In many cases, the vehicle can 102 be driven by an internal combustion engine. Another possibility is the vehicle 102 a hybrid electric vehicle (HEV) powered by both an internal combustion engine and one or more electric motors, such as a series hybrid electric vehicle (SHEV) Parallel hybrid electric vehicle (PHEV) or a parallel / series hybrid electric vehicle (PSHEV). Because the type and configuration of the vehicle 102 can also vary according to the capabilities of the vehicle 102 vary. As some other options, the vehicles can 102 different abilities in terms of passenger capacity, hitch and load and storage space.

The VCS 106 may be configured to support voice command and BLUETOOTH interfaces to the driver and RVs, to receive user input via various buttons or other controls, and to a driver or other occupant of the vehicle 102 Provide vehicle status information. An exemplary VCS 106 may be the SYNC ^® system, which is provided by the Ford Motor Company of Dearborn, Michigan.

The VCS 106 may also include various types of computing devices to support the performance of the functions described herein VCS 106 include. In one example, this can VCS 106 one or more processors adapted to execute computer instructions and a storage medium on which the computer-executable instructions and / or data may be managed. A computer-readable storage medium (also referred to as a processor-readable medium or storage) includes any nonvolatile (eg, physical) medium that participates in the provision of data (e.g., instructions) that may be provided by a computer (e.g. from the processor (s)). In general, a processor receives instructions and / or data, e.g. From memory, etc., to a memory and executes the instructions using the data, thereby performing one or more processes, including one or more of the processes described herein. Computer-executable instructions may be compiled or evaluated by computer programs constructed using a variety of programming languages and / or technologies including, but not limited to, and alone or in combination Java, C, C ++, C #, Fortran, Pascal, Visual Basic, Python, Java Script, Perl, PL / SQL, etc.

The VCS 106 can be configured with the TCU 120A to communicate. The TCU 120A Can a variety of modems 122 which are capable of packet-switched or circuit-switched signaling of the situation. The TCU 120A can the operation of the modems 122 control such that an appropriate communication path is used. The modems may be configured to communicate over a variety of communication paths. The paths can be circuit-switched 130 , packet-switched 132 . 134 Signaling or combinations thereof be configured. The ways of packet-switched communication 132 . 134 may be based on Internet Protocol (IP) or use packet-based arbitration to transmit information. For example, packet-switched communication may be long-term evolution (LTE) communications. In some cases, the route may be circuit-switched 130 Communication SIGTRAN or another implementation that carries circuit-switched signaling information over IP. However, the underlying signaling information is still formatted under the circuit switched protocol.

The VCS 106 can also input controls 108 receive human-machine interfaces (HMIs) configured to interact with the vehicle 102 provide. For example, that can VCS 106 with one or more keys or other HMI controllers 108 that are configured to work on the VSC 106 (eg audio buttons on the steering wheel, a PTT button, controls on the dashboard, etc.), establish an interface. The VCS 106 can also have one or more ads 110 that are configured to To provide, drive, or otherwise communicate with a visual output to vehicle occupants via a video controller. In some cases, the ad can 110 a touchscreen, which is further configured to receive user touch-based inputs via the video controller, whereas the display 110 in other cases, can only be an ad without touch-based input capabilities. In one example, the ad 110 a head unit display located in a center console area of the cabin of the vehicle 102 is included. In another example, the display may 110 a screen of a combination instrument of the vehicle 102 his.

The VCS 106 may also be configured via one or more in-vehicle networks 112 or vehicle buses 112 with other components of the vehicle 102 to communicate. The in-vehicle networks 112 For example, one or more of the following may include: a local network of the vehicle controller ( CAN ), an Ethernet network or a media-based system transmission (MOST). The in-vehicle networks 112 can it VCS 106 enable with other systems of the vehicle 102 to communicate, such as a vehicle modem of TCU 120A (which may not be present in some configurations), a module 120B a global positioning system (global positioning system) GPS ) configured to provide information about the current location and direction of the vehicle 102 and various other vehicle ECUs configured to VCS 106 together. Among other things, the vehicle ECUs may include, for example, a powertrain control module (PCM). 120C configured to control engine operating components (eg, idle controllers, fueling components, pollution emission monitoring components, etc.) and to monitor engine operating components (eg, engine diagnostic status); a body control module (Body Control Module - BCM ) 120D , which is configured to manage various power control functions, such as exterior lighting, interior lighting, keyless entry, remote startup, and the status of access points (eg, closure status of the hood, doors, and / or trunk of the vehicle) 102 ); a radio transceiver module (RCM) 120E , which is configured to work with key fobs or other local devices of the vehicle 102 communicating; a module for Climate Control Management (CCM) 120F configured to control and monitor the heating and cooling system components (eg, compressor clutch and blower fan control, temperature sensor information, etc.); and a Battery Control Module (BACM) 120G which is configured to the charging status or other parameters of the battery 104 of the vehicle 102 to monitor.

In one example, this can VCS 106 be configured to communicate with the TCU 120A via a vehicle bus 112 on the communication features of the TCU 120A access. As some examples, the vehicle bus 112 a Controller Area Network (CAN) bus, an Ethernet bus or a MOST bus. In other examples, the VCS 106 with the server 150 via a server modem 152 using the communication services of the modems 122 communicate.

With reference to 2 can the vehicle 102 an internal combustion engine 113 , a startup generator 114 , a battery 116 and electrical consumers 118 include. The control network 112 can with all of these vehicle systems through sensors (eg fuel level sensor 115 , Oil sensor 117 ) or vehicle system controls (eg 120A . 120B . 120C . 120D . 120E . 120F 120G ) connected. For example, the control network 112 control the vehicle systems to provide autonomous control. The internal combustion engine 113 can be a direct mechanical connection to the starter generator 114 respectively. The starter generator 114 can with the battery 116 and the electrical consumers 118 be electrically connected. The battery 116 can with the electrical consumers 118 be connected. In response to listening to the above-mentioned conversation, the VCS 106 Recognize that the vehicle occupants want to drive to Detroit, if they have enough fuel. The VCS 106 can obtain data from vehicle sensors (ie, fuel level sensor 115 ) to determine the remaining fuel in the fuel tank. The VCS 106 can estimate the expected fuel consumption for the current location of the vehicle 102 request to the symphony orchestra. In fact, the vehicle can 102 Listen to the conversation of the inmates and provide a response when prompted, without it being necessary to reconsider the question or provide additional information.

Regarding the 3A-B is an algorithm 300 shown. The algorithm 300 starts in step 302 , An implementation of the algorithm 300 may include additional or fewer steps, and the steps may be performed in a different order. The steps may also be performed simultaneously, at similar times, or in sequence. In step 304 collects the VCS 106 or other processors collect verbal utterances. The verbal utterances may be words, statements, uttered words, or conversations from the occupants in or near the vehicle that are detectable by the microphone or an array of microphones 124 Are available. In step 306 the topics are determined within the verbal utterances. The topics can be determined based on any natural language processing algorithm. Any part of the language - one or a combination thereof - can be used to determine the topics (eg nouns, verbs). The topics are determined to be later associated with the question asked. The topics can be parts of a sentence or an entire sentence. The topics can be formed by verb-noun associations or other grammatical, syntactic or semantic associations.

In step 308 wait that VCS 106 to a recognition of a reference object. A reference object may be any word or phrase associated with the vehicle (eg, manufacturer, model, software vendor, infotainment system vendor, brand associated with the vehicle or vehicle manufacturer). The reference object may be part of a confirmation question formulation or a confirmation question.

In step 324 a context evaluation is compared with a syntax evaluation. The contextual evaluation may be based on the topic evaluation of the step in 306 based on established topics. This means that the confidence that a particular statement, phrase, or word is associated with a particular theme and context can be determined based on a variety of methods for learning natural language and machine learning. For example, the statement "the Detroit Symphony Orchestra plays tonight in Detroit" includes two themes. One topic in the statement may be the Detroit Symphony Orchestra. A second topic may be the city of Detroit. The context of the statement can be musical events. Since the statement is directly associated with the context, the context confidence score can be considered high, qualitative, or about 75 , quantitatively. Similarly, the statement "The Detroit Art Gallery is hosting an exhibition in Detroit tonight" includes two similar themes that may have some relation to the context of musical events. The contextual evaluation in this situation can be significantly less, since the theme is not concrete musical performance, because instead it refers to artistic performances. For this reason, the context confidence rating may be considered mediocre, qualitative, or about 50 , quantitatively. The highest context score can be compared to the syntax score.

The syntactic can syntactically determine if a confirmation question formulation exists and the reference object is part of the confirmation question formulation. For example, the confirmation question formulation may simply be "Ford®?" Even though Ford® stands at the end of a statement as a confirmation question formulation - indicating a backward cadence - it may also be at the beginning of a forward cadence. The algorithm may determine, using known and unknown methods, whether the reference object Ford® is syntactically at the beginning or end of the question. The confidence score as to whether a forward or backward cadence is being used may be qualitatively or quantitatively characterized as described above. A phrase with the highest syntax rating can be: "_ (topic) ___. What do you think Ford®? "While a formulation with a low rating may read," Hey Ford® _ (subject) _. "The reason for this is that a reverse cadence almost never ends with" Hey Ford® "in front of the Topic, and a phrase like "What do you think, Ford®?" Is ambiguous for whether the topic is in front of or behind the wording. For this reason, the syntax score is an indication of whether the cadence is forward or backward.

In step 326 and step 328 Non-query utterances are filtered by reference objects to prevent inadvertent requests. If the reference object is not part of a confirmation question formulation (eg, backward cadence) or an active prompt (eg, forward cadence), the algorithm returns to step 306 back. If the referral object is part of the confirmation question formulation according to a syntactic confidence score above a predetermined threshold - or above a threshold set relative to the context score - the algorithm goes to step 338 about. If the referral object is not part of the confirmation query formulation according to a syntactic confidence score above a predetermined threshold - or above a threshold set relative to the context score - the algorithm goes to step 328 about.

After a confirmation question formulation in step 326 has been detected, the algorithm selects an already-captured topic from the verbal utterances based on contexts, as in the sub-algorithm A 310 Are defined. The subalgorithm A 310 collects information to define contexts. Contexts can be categories of topics or other logical representations configured to represent classes of vehicle operating parameters. As in step 312 shown, determines the sub-algorithm A 310 Contexts within verbal utterances. Contextual determination may generally include nutritional information, while fact-finding is more focused on a question about nutrients in a candy bar. The list of contexts (such as nutrients, distance to the destination, places of interest, weather) can be highly responsive to the vehicle in step 314 so that generic information requirements are not available (eg answers to arithmetic, emphasis on words). This means that comprehensive question polling capabilities may be constrained by the manufacturer or occupants, assuming that questions related to the vehicle or the ride are present.

Similarly, in step 316 Vehicle operating parameters analyzed to provide contexts. For example, the vehicle may be configured to provide vehicle-specific parameter contexts that are available to the question and answer service. This means that contexts related to oil life, fuel level, state of charge, air conditioning status, engine temperature, or other vehicle parameters are determined by the contexts in step 316 can be made available. A machine algorithm or manufacturer may step in 318 select the operating parameters needed to retrieve the answer Are available. For example, the oil temperature may be an available vehicle parameter, however, a machine learning algorithm may determine that the context should not be provided because questions about engine oil temperature are rarely provided. Vehicle parameter contexts may receive a stronger weighting than the contexts of the linguistic utterance.

In step 320 the contexts are presented to the user. This means that the user can further select which contexts he would like answered by the answering service. The contexts can be made using the HMI controllers 108 or the display screen 110 being represented. The contexts can be read by the system to the occupants and the occupants can provide the correct context selection. For example, the vehicle may indicate: "Line 1: Weather." The occupant may then verbally confirm that line one is the correct choice by saying "one." The contexts can be displayed in such a way that the most frequently used contexts are displayed first. The contexts may also be presented in an order based on the verbal utterances already received and the frequency of the contexts to be discussed. For example, if the weather in Detroit is an issue that has been discussed during a trip, the weather context may be presented to the user as a primary option. In step 322 The context selections are made by the user for use in step 338 receive. The contexts can be assigned weights to improve the response service. For example, the heavily discussed weather context may receive a heavier weighting than the rarely discussed oil temperature. While all other topics are the same, the more weighted context receives a topic selection priority over the unweighted or less weighted context in the topic selection process of step 338 ,

In step 338 the selected contexts are known together with the identified topics. The algorithm can then recognize the topic to be determined based on the contexts. The topic selection can take into account the weights applied to the contexts in which each topic resides. For example, topics within the weather context may be given a priority over topics in the oil temperature context. Furthermore, the syntactic and semantic strength of the topics can be weighted. For example, a confidence score of the topic may be determined based on the condition of the verbal utterance. This means that formulations that have a grammatical match fall into the more weighted context, but are disregarded due to syntactic or semantic evaluation. In addition, proximity to the confirmation question formulation can be used to further weight the weight of the topic. For example, the rating of topics that are directly in front of the confirmation question formulation can be doubled or multiplied by a factor. This means that a context with a low weighting can be selected over a context with a high weighting if the topic is just before the confirmation question formulation and the syntactic and semantic score for the weather topic is low.

In step 340 becomes the selected topic with the highest trust rating for the server 150 sent to be answered. The server 150 Provides the most likely answer through the statistical and machine learning algorithms in it. Any answering service can provide the answer and the answer does not have to relate to a vehicle. The answer can be sent back to the vehicle and the occupants in step 342 being represented. The vehicle 102 can then in step 344 automatically operate the vehicle based on the answer or prompt the user to select a course of action. For example, if the selected topic was "I say we have enough fuel to get to Detroit," the vehicle can prepare for the symphony and drive the car autonomously to the destination.

In step 328 that determines VCS 106 whether the referenced object is part of a forward cadence. The determination of the forward cadence may be based on the comparison of contextual and syntactic evaluations. For example, the syntactic score may be high if accompanied by a greeting (eg, "Hey Ford®"). If the referral object in step 328 is associated with a forward cadence, ignoring that VCS 106 all utterances before the call and saves all utterances after the call in step 330 , In step 332 send that VCS 106 the query request determined from the stored verbal utterances to the server 150 , The server 150 Provides the most likely answer through the statistical and machine learning algorithms in it. Any answering service can provide the answer and the answer does not have to relate to a vehicle. The answer can be sent back to the vehicle and the occupants in step 334 being represented. The vehicle 102 can then in step 336 automatically operate the vehicle based on the answer or prompt the user to select a course of action. If the statement read, "Hey Ford®, we have enough fuel to get to Detroit," the vehicle can An affirmative answer prepare a route to Detroit to the symphony and navigate the car autonomously to the destination.

The terms used in the specification are words of description rather than limitation, and it is understood that various changes may be made without departing from the spirit and scope of the disclosure. As previously described, the features of various embodiments may be combined to form further embodiments of the invention, which may not be expressly described or illustrated. For example, an always-listening algorithm may be used to identify topics that require a response service and automatically provide a response to the topic, which may not be based on an operating parameter of the vehicle. While various embodiments may have been described as advantageous or preferred over other embodiments or prior art implementations with respect to one or more desired properties, one of ordinary skill in the art will recognize that one or more features or characteristics may be compromised in order to avoid such desired overall attributes of the system, depending on the specific application and implementation. These attributes may include, but are not limited to, cost, strength, life, life cycle cost, marketability, appearance, packaging, size, operability, weight, manufacturability, ease of assembly, and so forth. Accordingly, embodiments that are described as less desirable than other embodiments or prior art implementations with respect to one or more features are not outside the scope of the disclosure and may be desirable for particular applications.

According to the present invention, there is provided a vehicle comprising a controller configured to respond in response to an input originating from verbal utterances defining a group of phrases including a preceding and following topic having a reference object therebetween only one of the themes is syntactically selected to select one of the topics for generating a response to a query embedded in the group based on an operating parameter of the vehicle and the syntax, and to operate an interface to output the response ,

According to one embodiment, the contexts are generated based on statements made while a response algorithm is not invoked.

According to one embodiment, the previous and following topics are determined based on the contexts.

According to one embodiment, a context score of at least one of the contexts and a syntax score are compared to determine which topic the referral object is associated with.

According to one embodiment, a context of the operating parameter and a context of the verbal utterances are the same.

In one embodiment, the operating parameter is a location of the vehicle.

In one embodiment, the operating parameter is the remaining oil life.

In one embodiment, the operating parameter is a fuel level of the vehicle.

In one embodiment, the response is based on an estimate of the fuel level for removal to an empty tank.

In one embodiment, the controller is further configured to operate a vehicle control system based on the response such that the fuel level is above zero until a destination is reached.

According to one embodiment, the reference object is a manufacturer of the vehicle.

In one embodiment, the controller is further configured to operate a vehicle control system based on the response.

In one embodiment, the vehicle control system is operated to ensure that the response is satisfactory based on the operating parameter.

According to one embodiment, the contexts are generated based on suction while a response algorithm is not called to determine the topics.

According to one embodiment, the contexts are selected by a user.

In one embodiment, the response is retrieved when the topic is discovered.

According to the present invention, there is provided a vehicle comprising a controller configured to respond in response to an input originating from utterances defining a group of phrases including a preceding and a subsequent topic having a reference object therebetween only one of the topics is syntactically selected to select one of the topics for generating a response to a question embedded in the group based on an operating parameter of the vehicle and the syntax, and operate the vehicle according to the answer.

According to one embodiment, the contexts are generated based on statements made while a response algorithm is not invoked.

According to one embodiment, the previous topic and the following topic are determined based on the contexts.

In accordance with the present invention, a method, in response to an input derived from utterances that define a group of phrases, including a previous topic and a subsequent topic having a reference object therebetween associated with only one of the topics by syntax, includes; Generating a response to a question embedded in a group of topics based on a selection of one of the topics and an operating parameter of the vehicle and the syntax, and operating the vehicle in accordance with the response by a controller.

Claims (15)

Vehicle comprising: a controller configured to respond in response to an input resulting from verbal utterances defining a group of phrases, including a preceding and following topic having a reference object therebetween associated with only one of the topics by syntax select the topics for generating a response to a question embedded in the group based on an operating parameter of the vehicle and the syntax, and operate an interface to output the answer. Vehicle after Claim 1 where the contexts are generated based on statements made while a response algorithm is not invoked. Vehicle after Claim 2 , whereby the previous topic and the following topic are determined on the basis of the contexts. Vehicle after Claim 3 in which a contextual evaluation of at least one of the contexts and a syntax score are compared to determine which topic the referenced object is associated with. Vehicle after Claim 1 where a context of the operating parameter and a context of the verbal utterances are the same. Vehicle after Claim 1 wherein the operating parameter is a location of the vehicle. Vehicle after Claim 1 , where the operating parameter is a residual oil life. Vehicle after Claim 1 , wherein the operating parameter is a fuel level of the vehicle. Vehicle after Claim 8 where the answer is based on an estimate of the fuel level for removal up to an empty tank. Vehicle after Claim 5 wherein the controller is further configured to operate a vehicle control system based on the response such that the fuel level is above zero until a destination is reached. Vehicle after Claim 1 , wherein the reference object is a manufacturer of the vehicle. Vehicle after Claim 1 The controller is further configured to operate a vehicle control system based on the response. Vehicle after Claim 8 wherein the vehicle control system is operated to ensure that the response is satisfactory based on the operating parameter. Vehicle after Claim 1 where the contexts are generated based on sucking, while a response algorithm is not called to determine the topics. Vehicle after Claim 10 , where the contexts are selected by a user.

Images